Authors:
Francesco Colace
1
;
Massimo De Santo
1
;
Mario Vento
1
and
Pasquale Foggia
2
Affiliations:
1
DIIIE, Università degli Studi di Salerno, Italy
;
2
DIS, Università di Napoli “Federico II”, Italy
Keyword(s):
Bayesian Networks, MultiExpert System
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Artificial Intelligence and Decision Support Systems
;
Bayesian Networks
;
Biomedical Engineering
;
Data Engineering
;
Enterprise Information Systems
;
Health Information Systems
;
Information Systems Analysis and Specification
;
Knowledge Management
;
Ontologies and the Semantic Web
;
Society, e-Business and e-Government
;
Soft Computing
;
Web Information Systems and Technologies
Abstract:
The determination of Bayesian network structure, especially in the case of large domains, can be complex, time consuming and imprecise. Therefore, in the last years, the interest of the scientific community in learning Bayesian network structure from data is increasing. This interest is motivated by the fact that many techniques or disciplines, as data mining, text categorization, ontology building, can take advantage from structural learning. In literature we can find many structural learning algorithms but none of them provides good results in every case or dataset. In this paper we introduce a method for structural learning of Bayesian networks based on a multiexpert approach. Our method combines the outputs of five structural learning algorithms according to a majority vote combining rule. The combined approach shows a performance that is better than any single algorithm. We present an experimental validation of our algorithm on a set of “de facto” standard networks, measuring pe
rformance both in terms of the network topological reconstruction and of the correct orientation of the obtained arcs.
(More)